Zero-Sum Game: After Costs, Negative

Michael Whitman · Trading since 2017, full-time since 2020 · Published 2026-09-06 · Updated 2026-10-19 · Reviewed 2026-10-03

How it works

A zero-sum game has a fixed total. Every dollar gained is a dollar somebody else lost, and no value is created or destroyed by the activity itself.

Share ownership is not zero-sum. The companies underneath produce goods, earn profits and pay dividends, so all the holders can gain together over time without anybody losing.

Derivatives are different. A futures contract or an option is an agreement between two parties, so every dollar one side gains is a dollar the other side lost. No production happens anywhere in the transaction.

Costs make it negative

The intermediaries are paid regardless of outcome. Spreads, commissions and exchange fees come out of the pool before it is distributed, which means the participants as a group finish below where they started.

So the average result across participants is a loss equal to the costs. Not because anyone is unskilled; the group as a whole has paid them. That says nothing about how many individuals lose, because a few large wins can sit beside many small losses.

Which is where a trading profit has to come from. In a growing business a return can come from the growth; inside a contract it can only come from the other party, and above costs it can only come from the other party doing worse than you.

A worked example

Say two traders take opposite sides of one futures contract and each pays $5 for the round trip. Whatever one wins the other loses, so before costs the pair nets to zero. After costs the pair ends $10 poorer between them, and the $10 went to the broker and the exchange.

Now put it in the CFTC’s own numbers. In the 29 September 2026 report, corn futures had 1,857,317 contracts of open interest. Commercial traders were net short 448,508 contracts, non-commercial traders were net long 509,501, and the small traders below the reporting level were net short 60,993. Those three net positions add to exactly zero: every contract bought by one group was sold by another.

Not every counterparty is competing

Some participants are buying insurance rather than seeking profit. A producer selling futures to fix a price is transferring risk, and losing on the contract while the underlying business gains is an acceptable outcome for them. That is what hedging means.

The CFTC sorts reportable traders on exactly this line. Its explanatory notes, read on 3 October 2026, say a trader’s positions in a commodity are classified as commercial if the trader uses futures in that commodity for hedging as defined in CFTC Regulation 1.3; everyone else who is large enough to report is non-commercial. Knowing which kind of counterparty you face explains more about the economics than any view about direction, and it is why commodity futures behave differently from a purely speculative contract.

The market maker is a third kind. A firm quoting both sides earns the spread and aims to hold no position at all, so its profit does not require you to be wrong; it requires you to transact. Speculators are competing with you, hedgers may be paying you to carry risk, and market makers are charging you.

The original data

The CFTC’s notes state the accounting behind the zero sum in one line: the aggregate of all long open interest is equal to the aggregate of all short open interest. Its 2026 legacy futures-only history file, downloaded on 3 October 2026, holds 14,055 market-weeks for 431 markets (433 market names, as two were renamed during the year) from 6 January to 29 September 2026. In 13,995 of them, the long positions of every group add to the total open interest, and so do the short positions. The other 60 rows are all three combined stock-index lines (S&P 500, Nasdaq-100 and Dow consolidated), where the long or short total misses the open interest by 1 or 2 contracts.

Who is on the other side depends on the market. In the 29 September 2026 report, the share of each side of open interest held by commercial traders was:

Gold shows the hedger pattern most plainly. Commercials held 76.22% of the short side and 14.47% of the long side, so the open short positions a gold buyer was lined up against were mostly commercial ones. In the E-mini S&P 500 and the 10-year note, commercials held most of both sides. The label is looser in financial futures: the CFTC’s notes say one financial organization may have a banking entity classed as commercial and a money-management entity classed as non-commercial. The eight-market table is published with the raw contract counts.

Week by week, one group’s net position mirrors the others’. In corn, commercials were net short in 32 of the 39 weekly reports of 2026, and their net ran from 476,845 contracts short to 68,044 long while the non-commercial net ran from 51,704 short to 542,999 long. In the E-mini S&P 500, commercials were net short in 20 of 39 weeks. In every one of the 78 rows of the weekly net-position table, the three nets sum to zero.

The share side looks nothing like this. Yahoo Finance’s SPY history, downloaded the same day, lists 136 dividends from 19 March 1993 to 18 September 2026, adding to $113.38 a share, $7.28 of it in the four payments of 2025. Over the same span the price rose from $43.94 to $769.64. None of that dividend money came from a counterparty losing a trade; it came out of company earnings, which is the difference between owning a business and holding a contract on its price.

In practice

Volume is the size of the field. Heavily traded markets are efficient because a great many people are looking for the same mistakes, which makes an edge harder to find and cheaper to trade when found.

Lengthening the horizon changes which game you are in. Owning productive assets for years is positive-sum; trading contracts against other participants for minutes is not, and the choice between them is a choice of game rather than of skill.

A gap transfers the entire move at once. Somebody’s loss on that print is somebody else’s gain, and neither had an opportunity to act.

Every stop that fills meets an order on the other side. That is not a conspiracy, it is what a transaction is, and it is worth remembering when a stop-out feels targeted.

How this connects

Payment for order flow shows where part of the cost goes. A broker routing orders to a wholesaler is paid for that flow, one of the ways the pool’s leakage reaches intermediaries.

The clearing house stands between the two sides of an exchange-traded contract, which is why the winner’s gain and the loser’s loss pass through one ledger.

Perpetual futures make the transfer visible. A funding payment passes between longs and shorts at set intervals, a direct zero-sum payment on top of the price move.

Short-term forex speculation sits in the same category. Apart from interest differentials, a currency produces nothing, so gains come from other participants, less the costs.

Day trading lives squarely inside the negative-sum version, and slippage is one more cost leaving the pool that never appears on a fee schedule.

And the positive-sum alternative is an index fund held with buy and hold, where company earnings and compound interest do the work instead of somebody else’s mistake.

When it fails as a framing

In a range the description is most literally true. Nothing is produced, prices end where they began, and the only money that moved was the costs, which is the clearest illustration of what a negative-sum game actually is.

The second misreading is applying it to shares. Long-term equity ownership is positive-sum, and importing this framing there produces unnecessary pessimism.

A third is concluding that somebody must win. After costs, every participant in a period can finish behind; the arithmetic only fixes the total, not who holds it.

A fourth is ignoring the hedgers. A commercial that loses on a contract may have gained on the crop or the metal it was hedging, so a futures loss is not always a loss to the business.

And a fifth is forgetting the costs. They are the reason the game is negative rather than balanced, and they are the one part fully under your control.

Related

Futures is the clearest example of the structure. Options is the other, with an extra dimension. And why traders lose money is where this arithmetic shows up in outcomes.

This page is educational, not financial advice. Test every idea on your own charts before risking money.